What's happened?

My house move completed yesterday with the line being connected and fibre also. Initially Plusnet had 'accidentally' provisioned me on the unlimited package despite me being on the unlimited extra in my previous house. When first connected yesterday, these were my stats:

Downstream Upstream
Line Coding (Trellis) On On
SNR Margin (dB) 18.5 20.0
Attenuation (dB) 19.5 0.0
Output Power (dBm) 12.4 4.0
Attainable Rate (Kbps) 84485 16190
Rate (Kbps) 39998 1999

Now, after 2 disconnects (not of my doing!!) overnight, they are:

Downstream Upstream
Line Coding (Trellis) On On
SNR Margin (dB) 6.1 6.3
Attenuation (dB) 19.6 0.0
Output Power (dBm) 12.6 7.6
Attainable Rate (Kbps) 57039 16078
Rate (Kbps) 57031 16112

Obviously I'm not super impressed with my download speed. My attenuation looks all wrong seeing as I'm only about 147m from the cab.

My previous house, despite being only 5 door away was about 500m from the cab due to line being routed differently via an SCP but I managed to get a constant 65Mbps there with an attenuation of 18!

No point contacting Plusnet yet as they'll tell me they can't do anything until after the mythical '10 day training period' but anyone got any ideas why my attenuation is so high and snr so low?

So the figures previously were an anomalous on the Downstream ...

Those now all tie together reasonably nicely.

Can you be sure you are only 147 metres away ... the attenuation is more consistent with a 600m line length. Where did the 147 come from?

An Openreach engineer I was talking to next to the cab last week. Also, my neighbours either side and across the road all get rates in the late 60s to early 70s. There is still noise on the line in the background, could this be causing it?

Edited by WelshWArrior (Thu 18-Jun-15 08:26:52)

How did he come up with the figure?


Given the attenuation has remained the same with the resets, I would guess it is accurate.

I think he was working on the house directly opposite as I saw an Openreach van parked outside about 20mins later.

Could the noisy line cause low snr and high attenuation?

Just did a check on the availability for neighbours either side of me with their phone numbers, I got this:

High Low High Low
FTTC Range A (Clean) 80 80 20 20 -- Available
FTTC Range B (Impacted) 80 61.9 20 12.2 -- Available

Which is more in line with the stats I saw when first connected yesterday.

Edited by WelshWArrior (Thu 18-Jun-15 08:50:55)

Did he actually do any measurements on your line?

Noise could cause low SNR however does not affect attenuation. Attenuation is mainly due to the line length and gauge of wire with very minor effects caused by the actual termination impedance of the modem and a some other factors.

Hmm something strange is going on. I just spoke to next door neighbour and he confirmed he gets a constant 74meg and he's further away from the cab than me. Also, the checker results I posted above for the neighbour the other side suggest that I should be getting a hell of a lot more download speed don't they?

Engineer didn't do check on my line - I just asked him how long it was.

Edited by WelshWArrior (Thu 18-Jun-15 09:22:14)

Looks like something is wrong based on the BTw check results...

Plusnet will look at getting the voice fault fixed first before anything else...

In reply to a post by WelshWArrior:

Engineer didn't do check on my line - I just asked him how long it was.

So, he just plucked a number out of the air. Do you really think he knows the length of every line?

Of course he doesn't and to be honest, I don't care what he said the line length is. What I DO care about is that the houses either side of me are getting late 60s early 70s down (which is supported by the checker results I posted for my next door neighbour) which means so should I.

I can literally throw a stone and hit the cab. There are BT manholes outside each house coming up from the cab and I know the cable comes straight up the hill past my house.

I just measured on Google Earth and I know it's only rough but got 163m from the cab to where the drop cable comes into my house.

Just because you neighbour can get 70Mbps does NOT mean you can too.

The point is that your attenuation is at a level which would deliver high fifties or maybe just over sixty. Without knowing the actual cable length it is difficult to confirm if the attenuation is correct. There may be an access point outside your house and you "think" you know the cable comes straight to you, however it could go to a property further away, have their pair(s) taken out before coming back to you.

I think we need to forget about the cabling and look at the fact that BT Wholesale checker says he should get 80/20 and 80/20 on impacted line....

If it was self install it says lowest would be 61/12 but this isn't the case...

Where are those actual figures? They are not posted in this thread.

Is there a way to tell how long the line actually is? Would it be in the Engineer's installation report and would Plusnet have access to that?

When the Engineer rang me yesterday morning to ask if he could possibly do my installation 4hrs early (!!), he said that he'd already tested the line and found a fault that he was going to attempt to fix - could this be causing a problem which would drop the attenuation? Assuming he didn't fix it of course - which the noise on the line would suggest.

Yes they are

Finally some sense

Is it worth mentioning it to Plusnet or are they likely to give me the usual speech of having to wait until the 10 days is up? The line noise make me think they'll probably listen to me.

His test equipment has a Time Domain Reflectometer (TDR) function which sends pulses down the line and measures the time taken for a reflection to be returned. Similar to you shouting and waiting for an echo.

By measuring the time taken - on a 500 m line it will be around 3.6 microseconds, the line length can be measured to a good degree of accuracy. Any faults such a corroded joints or factors such as bridge taps will also be shown and the distance of those provided.

Faults will normally impact in one of two ways - very minor increases in attenuation or a very large increase. A single leg disconnected - however as voice works, that is not a problem. An extra termination on the line - possible but unlikely.

Why not put the original modem back and see if the fault you detected is still there or if it has been fixed.

The SNRM of 6.1 isn't low. It's standard.

Your actual and attainable are now right in line with what I said in your previous thread regarding the attenuation.

You mention noise on the line. That probably would affect speed, but I'm not sure what kind of fault would cause inflated attenuation. Did your engineer confirm the nearby cab is the one you're on? We sometimes see ex-showhouses on one further away, the nearby one being added when the estate is well into the build stages.

Noise on the line isn't subject to the 10-day myth. I suggest you give Plusnet a kick so we eliminate that first. Was the house previously on FTTC do you know? The OR VDSL2 filter would have given that away if present when you bought it, or more surely when you took over - before the line was activated.

Care to point to the OPs actual figures?

You said the noise went away when using the Billion. Has it come back?

If not, that suggests it was the modem (or PN router?) power supply possibly dodgy.

In reply to a post by mlmclaren:

I think we need to forget about the cabling ....

That's what we do need to look at. His attainable and actual are spot on for the attenuation. So unless something is causing that to be high he isn't going to get much more.

Hi Roberto

It's got to be that cab as it's the only one live within a mile. It was an old lady that lived here before we bought it so no VDSL plate and I'm pretty sure she probably didn't have Internet either. The line had been totally disconnected back in October after she died.

I think the best thing to do is put the Openreach modem back on when I pop home at Lunchtime and check if the noise is still there. If it is, I can report it to Plusnet and hopefully get an engineer back out. I still have the mobile number of Engineer that called yesterday - very tempted to give him a quick ring but better not

Yes it is back but not as pronounced.

In reply to a post by MHC:

Care to point to the OPs actual figures?

Yeeees http://forums.thinkbroadband.com/fibre/t/4414929-re-...

In reply to a post by WelshWArrior:

Finally some sense

Is it worth mentioning it to Plusnet or are they likely to give me the usual speech of having to wait until the 10 days is up? The line noise make me think they'll probably listen to me.

If you report noise on Phone then Plusnet "should" something about that immediately

Another thought .

Assuming you are still using the Billion, but the OR modem back on, (with a different ADSL cable if you haven't tried one yet), then after a couple of minutes run the BT Wholesale speed test (ignoring the red instructions as in effect you'd have just done them), and continue to the Further Diagnostics.

Tell us the downstream IP Profile from that. We can calculate back to see if it is still 57Mbps. The actual speed will still reflect the 57Mbps even if the IP Profile has gone up, due to the Plusnet Current Line speed lagging on change.

To be fair the MHC, the estimated quoted above is for my neighbours either side of me using their mumbers. Mine is as below:

Telephone Number 01834844xxx on Exchange TENBY is served by Cabinet 3
Featured Products

Downstream Line Rate(Mbps)

Upstream Line Rate(Mbps)

Downstream Range(Mbps)

Availability Date

High Low High Low
FTTC Range A (Clean) 59.6 39.3 16.1 10 -- Available
FTTC Range B (Impacted) 50.8 28.2 15.9 7 -- Available

BUT - and this is a big BUT, those are for my OLD house that I moved from which is 5 doors away but over 300m further in line length due to it going around corner and to SCP before continuing onto old house. The figures don't seem to have been updated since I moved last friday - they are identical.

Edited by WelshWArrior (Thu 18-Jun-15 11:30:43)

In reply to a post by mlmclaren:

In reply to a post by MHC:

Care to point to the OPs actual figures?

Yeeees http://forums.thinkbroadband.com/fibre/t/4414929-re-...

NO. Read the post:

Just did a check on the availability for neighbours either side of me with their phone numbers, I got this:

High Low High Low
FTTC Range A (Clean) 80 80 20 20 -- Available
FTTC Range B (Impacted) 80 61.9 20 12.2 -- Available

The OP has also said that the line to the property was disconnected last October so a check on the number could not have been made and any estimates would have been using the address checker which is acknowledged to be a very rough estimate.

"for neighbours either side of me with their phone numbers"

That makes sense that they haven't changed yet. In effect at the new address it is a new number and for a few days it always shows the results for wherever that number previously landed.

In reply to a post by RobertoS:

In reply to a post by mlmclaren:

I think we need to forget about the cabling ....

That's what we do need to look at. His attainable and actual are spot on for the attenuation. So unless something is causing that to be high he isn't going to get much more.

OK well who's popping round to have a look at the cable then??

Sorry for my sarcasm above but we have got to look at this from a point of reporting it to the provider to do something about the issue at hand...

Though to be fair the provider is first going to want the noise issue fixed... this could be the issue if there is a cable fault somewhere

But contacting Plusnet and blabbering on about the cable isn't going to get much response.

Edited by mlmclaren (Thu 18-Jun-15 11:50:17)

oh balls...

oh ok so what does the address checker for your new place give!

Another thought re the noise.

Before reporting it you need to check using the Quiet line test 17070 Option 2 with a (preferably corded) phone into the test socket. To get to the test socket you need to remove the VDSL2 filter plate, which could be faulty.

If there still is noise you have a good case. If there isn't, then see what you get through the phone socket of the VDSL2 filter plate before refitting the NTE5A faceplate, then after refitting it.

Do all extensions in the house stop working when the NTE5A faceplate is removed?

Plus, it's worth seeing what you get using a dangly filter into the test socket once the VDSL2 filter is off.

Presuming that is noise when all broadband hardware is disconnected.

He's going to have to wait for the checker database to catch up before he can post the correct information!

Not sure how long it will be, at least a few days I would have thought.

Noise when a modem is connected can mean

Dodgy filter
HR fault
Faulty modem

If the noise is present when all filters and broadband kit is removed and just a corded phone in the test socket, then you have a nice easy voice fault that needs tracing

Yep I'll do this at lunch as well as changing over from my billion 8800nl to the openreach modem.plusnet router.

There are 0 extensions in the house as I ripped the spiders web of wiring out. The master socket WAS in the diningg room some way from where the drop cable comes into house and went via all sorts of junction boxes. I got engineer to put new VDSL master socket right where drop cable enters house and modem is plugged straight in.

Good thought - I'll give that a go too.

In reply to a post by rippedcotton:

He's going to have to wait for the checker database to catch up before he can post the correct information!

It occurred to me a few minutes ago that the Openreach estimates database and the BT Wholesale feed from it now track actuals much quicker on FTTC than they do/did on ADSLx.

There's a nasty chance that by the time the BT Wholesale checker gets the new data for the number that it will reflect what the OP is getting .

In reply to a post by RobertoS:

In reply to a post by rippedcotton:

He's going to have to wait for the checker database to catch up before he can post the correct information!

It occurred to me a few minutes ago that the Openreach estimates database and the BT Wholesale feed from it now track actuals much quicker on FTTC than they do/did on ADSLx.

There's a nasty chance that by the time the BT Wholesale checker gets the new data for the number that it will reflect what the OP is getting .

It's possible, but that wouldn't explain the apparent discrepancy between his stats and what his neighbours' checker results suggest is possible. If it's really 150m from the cab then 80/20 should be as easy as falling off a log.

Makes it a lot harder for any ISP to get anyone in BT Wholesale or Openreach to listen, even if the ISP can be persuaded to try.

BTW and OR now resist spending money to fix poor quality lines, even when it's clear there is a problem.

Yet, as part of BDUK, I know they've definitely improved the copper (or metal, anyway) in 2 villages here ... In order to improve the FTTC speeds.

I've uploaded a pic of where the cab is (bottom circle) and where my house is and the route the cable takes according to an Openreach Engineer that installed my fibre a few months ago in the house we just sold a few doors up. Distance from cab to house according to Google Earth is roughly 163m. Pic here.

The houses either side get late 60s to early 70s down and the house directly opposite over the road gets 72Mbps according to the guy I spoke to outside last night.

1. Need to see actual speed test results, have come across people who quote sales estimates rather than actual tests
2. Need to compare the attenuation values from the various households to see if similar
3. If the attenuation is vastly different then suggests for some historical reason your line is a lot longer, or the phone does not work and only one leg (1 wire from the pair) is working.

I'm planning on seeing the neighbours either side this weekend and asking if I can connect to their modem and perform speedtest. I'll post the results.

It is not just the speedtests results that are needed. You need to get the basic connection stats too: Use DSLStats, under Telnet Data the second level Tab Connection Stats then either take a save the text using the camera ICON or clipboard to copy and paste, and maybe the data under the Attenuation Log tab.

In reply to a post by RobertoS:

Makes it a lot harder for any ISP to get anyone in BT Wholesale or Openreach to listen, even if the ISP can be persuaded to try.

BTW and OR now resist spending money to fix poor quality lines, even when it's clear there is a problem.

My experience is that while I was able to get things fixed in roughly 2005 (and it took a voice line fault to do that), a similar problem in 2009 was the very devil to get resolved. In fact if it hadn't have been that some ne'erdowell made off with 1/4 mile of E-side cabling which had to be replaced then it may never have been. Loss of service for 4 days was worth it in the end

I'll get as much as I can. I don't know the neighbours very having only move din last friday but I've got until next Saturday anyhow as Plusnet won't even look at this until after the mythical '10 day training period'!!

As several of us have said, the noise on the phone if present with a corded phone, (£5-£8 from lots of places), through the test socket is nothing at all to do with broadband, and the 10 days does not apply.

I have no phone plugged into the socket now as I'm going to leave it unplugged for a few days to see if that makes any difference at all - not expecting it to. Having said that, since unplugging it yesterday evening, my sync has risen from the 52Mbps it was at back up to 56Mbps.

I have also now ordered a high quality rj11-rj11 cable (1m in length) as mine is several years old and I noticed the cat has been chewing it so it may also be contributing to the Attenuation. I'm probably clutching at straws mind

Obviously my best bet it to talk the neighbours into letting me connect to their routers and get some stats along with some speed tests.

Edited by WelshWArrior (Fri 19-Jun-15 14:54:01)

In reply to a post by WelshWArrior:

BUT - and this is a big BUT, those are for my OLD house that I moved from which is 5 doors away but over 300m further in line length due to it going around corner and to SCP before continuing onto old house. The figures don't seem to have been updated since I moved last friday - they are identical.

Is there a reason to suspect that your new house's wiring goes a different way? It *could* go the same way.

In reply to a post by WelshWArrior:

I've uploaded a pic of where the cab is (bottom circle) and where my house is and the route the cable takes according to an Openreach Engineer that installed my fibre a few months ago in the house we just sold a few doors up. Distance from cab to house according to Google Earth is roughly 163m. Pic here.

The BT engineer probably gave you the "default" routing, without knowing what precise deviations have happened for individual homes.

There could have been a case, long in the past, where your "normal" DP ended up full, so, upon ordering, your line was routed to some other DP. From your description, it looks like this happened to your old home, so you shouldn't immediately discount it as a possibility for the new place.

The houses either side get late 60s to early 70s down and the house directly opposite over the road gets 72Mbps according to the guy I spoke to outside last night.

It is easy *now* to think of all homes being wired for phone lines in a consistent manner, as the infrastructure gets put in place during the build. Back in the fifties & sixties, this wasn't the case. It was easy for the infrastructure to run out of capacity, and for odd routing to occur.

Of course, you might be routed the shortest way, and the speeds are all limited by the voice fault. It really isn't worth trying to sort out *anything* with the broadband until an audible voice fault is fixed.

In reply to a post by WelshWArrior:

I have no phone plugged into the socket now as I'm going to leave it unplugged for a few days to see if that makes any difference at all - not expecting it to. Having said that, since unplugging it yesterday evening, my sync has risen from the 52Mbps it was at back up to 56Mbps.

I have also now ordered a high quality rj11-rj11 cable (1m in length) as mine is several years old and I noticed the cat has been chewing it so it may also be contributing to the Attenuation. I'm probably clutching at straws mind

Voice faults rarely fix themselves spontaneously, nor from leaving a phone unplugged.

Obviously my best bet it to talk the neighbours into letting me connect to their routers and get some stats along with some speed tests.

Getting some figures for comparison will help your understanding. But my bet is that it doesn't beat having an engineers attention focussed on your line.